Lung cancer is the leading cause of mortality worldwide. Although tobacco control is the best method for preventing this disease, there is little indication that smoking rates will decrease much further in the U.S. We documented that there exist 2-to-5-fold differences among US racial/ethnic minorities in the risk of lung cancer associated with cigarette smoking, even after taking into account self-reported dose (cigarettes per day) and duration. Native Hawaiians and African Americans have a significantly higher risk for lung cancer than European Americans, who in turn have a significantly higher risk than Japanese Americans and Latinos. We hypothesize that these differences are related to common genetic variants that affect internal dose and metabolism of nicotine and tobacco smoke carcinogens, and/or DNA repair. Our preliminary data obtained collaboratively with the laboratory investigators on this POI suggest that a combination of genetic and behavioral factors acting on several metabolic and/or DNA repair pathways are likely to explain the observed ethnic differences in risk. We propose to conduct a genome-wide association study (GWAS) to comprehensively assess the role of genetic variation in predicting 11 biomarker phenotypes (measured by Core B) related to tobacco carcinogen and nicotine exposure and metabolism among 2,250 Native Hawaiian, African American, European American, Latino and Japanese American smokers in the Multiethnic Cohort study, using stored biospecimens. We expect common inherited variation to assist in better assessing one's actual exposure level (as compared to cigarettes/day) and ability to metabolize tobacco carcinogens and, thus, in better predicting risk of developing lung cancer. We also expect the underlying causal variants to be more common in ethnic/racial populations at high risk for lung cancer. Finally, as part of this project, we will test the genetic variants associated with these phenotypes, as well as variants in DNA repair genes, for association with lung cancer incidence among 1,448 incident cases and 1,448 matched controls ofthe same five ethnic groups in the MEC. The results of these studies will provide important information on the biological determinants of internal dose and metabolism of nicotine and tobacco carcinogens and their relationships to lung cancer risk in U.S. ethnic/racial minorities.